Planning, Search, Certainty and Satisfaction Among Durables Buyers: a Longitudinal Study

Anthony Cox, Indiana University
Donald Granbois, Indiana University
John Summers, Indiana University
ABSTRACT - Preliminary results of a two-wave longitudinal study of durable goods purchase plans and fulfillment supported the value of further investigating four acquisition categories for major durables, including Maintenance Replacement, Adjusting/Upgrading Replacement, Additional Unit Expansion, and First Acquisition Expansion. These categories were more useful than the distinction between "planned" and "unplanned" measures in explaining variations in search, satisfaction and payment method variables.
[ to cite ]:
Anthony Cox, Donald Granbois, and John Summers (1983) ,"Planning, Search, Certainty and Satisfaction Among Durables Buyers: a Longitudinal Study", in NA - Advances in Consumer Research Volume 10, eds. Richard P. Bagozzi and Alice M. Tybout, Ann Abor, MI : Association for Consumer Research, Pages: 394-399.

Advances in Consumer Research Volume 10, 1983      Pages 394-399

PLANNING, SEARCH, CERTAINTY AND SATISFACTION AMONG DURABLES BUYERS: A LONGITUDINAL STUDY

Anthony Cox, Indiana University

Donald Granbois, Indiana University

John Summers, Indiana University

ABSTRACT -

Preliminary results of a two-wave longitudinal study of durable goods purchase plans and fulfillment supported the value of further investigating four acquisition categories for major durables, including Maintenance Replacement, Adjusting/Upgrading Replacement, Additional Unit Expansion, and First Acquisition Expansion. These categories were more useful than the distinction between "planned" and "unplanned" measures in explaining variations in search, satisfaction and payment method variables.

INTRODUCTION

The authors of a recent comprehensive review of research on consumer behavior for major household durables called for further research on the topic, arguing that:

In contrast to the repetitive purchasing of household detergents or toothpaste, consumer purchase behavior for durable goods is much more consequential and complex. Researching the subject is similarly more challenging and more exciting. Regrettably, academics have directed more attention to toothpaste and detergents than the inherently more interesting automobile, refrigerator, or microwave oven. (Dickson and Wilkie 1978, p. 151).

The present paper gives a preliminary and highly selective report on a project undertaken by several researchers who wholeheartedly share Dickson's and Wilkie's enthusiasm for the inherent interest and excitement offered by durable goods research. However, as our presentation of these results will abundantly demonstrate, we have found that their characterization of the research area as "challenging" is, if anything, an understatement. Nevertheless, we feel our experience with the project has given insights worth sharing.

Our intention in designing the portion of the study reported here was to help us understand the character of the factors initiating planning for a prospective durable purchase and the nature of the information search process rather than to build predictive models of the outcomes of the process. We feel such understanding has relevance both for the further development and refinement of theoretical models of consumer behavior and for understanding and influencing durable goods purchases in more practical settings.

IDEAL VS. ACTUAL PROCESSES

Normative writings in consumer and home economics and early expositions of consumer decision process models have portrayed the process of planning for and implementing major purchases as multi-staged and complex. In these writings, one or more triggering factors lead to problem recognition, which in turn initiates a sequence of planning behaviors in which priorities somehow are assigned to competing uses for household monetary, time and space resources; information from one or more sources is gathered and weighted, often in intra- and inter-family discussions; and current and future family income, saving and credit resources are estimated, all resulting in the purchase, use and subsequent evaluation of one or more major durables.

Reviews of the numerous studies of the planning, information search and shopping behaviors preceding actual durable good purchases, however, conclude that the somewhat idealized process described above does not always occur. Extraordinarily short planning periods are not uncommon, and substantial proportions of buyers report purchases in which a single brand was considered, only one store visited, and little or no external information search was undertaken. (Granbois 1977; Newman 1977; Olshavsky and Granbois 1979). Longitudinal studies, in which fulfillment rates for purchase plans reported earlier are determine:d in one or more follow-up waves, reveal substantial proportions of unfulfilled plans and surprisingly large numbers of purchases for which no plan was reported during the first wave interview (Dickson and Wilkie 1978, p. 96).

It seems clear that further conceptualization of the processes is required to bring about a closer fit between theory and research findings. Some very rudimentary elements of the conceptualization that guided our study are sketched out in the section that follows.

HOUSEHOLD DURABLES ASSORTMENT MANAGEMENT

Our view is developmental in that the acquisition and disposal of major household durables is seen as part of a large on-going process analogous to assortment management in a formal organization such as a business firm. The notion of assortment management was one of the many provocative ideas advanced by Wroe Alderson (1957), whose writings have influenced our approach.

The newly formed household starts out with a "starting set" of durables, that combination of large and small durables thought to be essential for supporting the desired activities of household members. At the minimum, the starting set includes those durables found by penetration studies to be available in virtually every household. Components of the starting set are acquired in one of these ways (note that items are not all necessarily "owned"):

1. already owned by one household member

2. provided with dwelling unit

3. received as a gift

4. purchased (new or used)

5. rented or leased

6. available for use elsewhere in building or complex

Beyond acquisition of the starting set, durable acquisitions may represent any of four types:

1. Maintenance Replacement, whereby an item in the existing assortment is replaced by another item roughly comparable in function and quality.

2. Adjusting/Upgrading Replacement, whereby an item in the existing assortment is replaced by another item that is functionally more suited to current household needs and/or higher in quality, style or prestige.

3. Additional Unit Expansion, whereby a unit is added to a category already represented in the household assortment. Such additions may be functionally similar to existing items (e.g., an additional window air-conditioning unit) or may represent an enhancement of the overall utility of the assortment by providing additional specialized functions (e.g., a station wagon added to an assortment already containing a sedan).

4. First Acquisition Expansion, whereby an item in a category not previously represented in the assortment is added. Such additions may be innovative products recently introduced into the market, or established products.

Except for studies in which households in the process of formation are an important part, most research on durable purchases includes purchase plans for and acquisitions of items representing a mix of the four categories outlined above. Through time, an individual household may make purchases in each of these categories, and at any one time may be considering a set of anticipated purchases representing more than one (or even all) of the four categories.

It seems intuitively obvious that considerable variation may occur among the four types of durable goods acquisition in the set of precipitating factors triggering problem recognition, the length of planning period, the degree of perceived risk (and the resulting degree of recognized need for new information), the sense of urgency felt by household members and their level of agreement as to the desirability and urgency of the proposed acquisition, the priority for funding assigned the project, and the susceptibility of the proposed purchase to both accelerating and disrupting effects of unexpected circumstances. A few of the consequences of these predicted differences are explored in the study results to be reported here. Later, more thorough analysis of our data will explore differences among the four categories in considerably more detail.

Our discussion here will be limited largely to five aspects of the shopping and purchasing process, including the number of stores visited, number of brands considered, payment method, whether respondents were certain at the time of purchase that the brand or model purchased was exactly what was wanted, and satisfaction with the item purchased. We will examine the general hypothesis that classifying durable goods purchases into the four acquisition categories defined earlier provides a more effective way to explain variance along these five dimensions than does the distinction between planned and unplanned purchases often made in longitudinal studies. We will also investigate the relationships between the "search" variables--number of stores visited and number of brands considered--and the certainty and satisfaction measures to test the common-sense prediction that greater search Yields higher certainty and greater satisfaction.

STUDY DESIGN

Two mail questionnaires were sent eight months apart to households in the University of South Carolina Consumer Panel. The first provided data on household durables inventories, purchase plans for major durables, and expected changes in housing and financial circumstances. To identify purchase plans, respondents were requested to list all items for which there was at least a .1 probability of purchase in the following 12 months. Any product meeting this criterion was considered to be "planned." For each product they listed, respondents indicated whether it was intended as a Maintenance Replacement, Adjusting/Upgrading Replacement, Additional Unit Expansion, or First Acquisition Expansion. The follow-up questionnaire asked details about purchases made and a short series of questions on household spending, saving and credit practices. Purchases listed on the follow-up questionnaire were identified either as "unplanned" on the first-wave questionnaire or as fulfillment of plans anticipated on that questionnaire. For all purchases, whether planned or unplanned, respondents indicated the number of brands considered, the number of stores visited, how certain they were at the time of purchase that the brand/model was what they really wanted, and their subsequent satisfaction with the item purchased.

The discrepancy between the 12-month planning period and the eight-month interval between questionnaire waves was deliberate. Juster (1966) found 12-month purchase probabilities predicted six-month purchase behavior better than did six-month purchase probabilities. Pickering and Isherwood (1974) judged a four-month interval too short and a 12-month interval too long in studies of this sort. They concluded, "There does not, however, appear to be any point in attempting to adhere rigidly in the collection of reinterview data to the time period to which purchase probabilities relate."

FINDINGS

Of the 800 South Carolina panel members included in the initial mailing, 566 returned acceptable questionnaires for both waves. The overall response rate of 71% was considered very acceptable given the large amount of data requested of each household. An examination of the demographics of the sample showed it to be older, more highly educated, and more established than the general population. The average age of the head of household was 49 years; 38% were college graduates; and 90% of the households owned homes. As such, durable purchases for the sample could be expected to include a substantial portion of replacement purchases for those durables with high levels of penetration.

Of particular interest in this study are purchases of 21 well-known household durables (e.g., kitchen and laundry appliances, home entertainment products, and motor vehicles). Table 1 presents the distribution of the number of durables purchased by individual household over the 8 month study period. The 566 households purchased a total of only 258 of these durables, which highlights the need for extremely large samples in durable goods research when sampling the general population. The number of durables purchased per household ranged from 0 to 5 with 387 households (68% of the sample) purchasing none of the 21 durables during this period. Of the 179 remaining households, 54 (30% of the purchase households) made multiple purchases and accounted for 51% (133 of 258) of the total purchases. The substantial number of multiple purchases reduces the validity of most types of statistical tests involving more than one durable since it violates the usual assumption of statistically independent observations.

TABLE 1

DISTRIBUTION OF NUMBER OF PURCHASES BY INDIVIDUAL HOUSEHOLDS

Table 2 presents an overview of the sample's plans and purchases. The number of households with plans ranged from a low of 1 for both stereo consoles and video tape recorders to a high of 137 for the household ' s first or primary automobile. Purchases (both planned and unplanned) varied from none for stereo consoles, video cape recorders and pianos to 60 (49 + 11) for first automobiles. Because for many durables there was only a small number of either plans or purchases (prohibiting the development of meaningful conclusions for these individual products), and since most seemed to fall. naturally into four basic product categories, the durables were so organized and product category totals calculated

The fulfillment rate was highest for vehicles (35%), a category dominated by first automobiles (137 of 173 plans), followed by entertainment appliances (27%), a category heavily weighted toward color televisions (78 of 120 plans). Fulfillment rates for kitchen appliances and for laundry appliances were substantially lower (17% and 14% respectively). One might speculate that the presumably greater excitement and entertainment value associated with the purchase of the products in the former two categories serves to motivate consumers more strongly to fulfill these plans.

The overall fulfillment rate (23%) was lower than those reported in seven previous studies, which found rates ranging between 312 and 42% (see Table 3). This is partially due to the use of an 8 month followup rather than the 12 month followup most frequently employed in these previous studies. Granbois and Willett (1968) found fulfillment rates to be approximately 1.17 times as high at 12 months compared with 8 months. This would suggest a predicted 12 month fulfillment rate of 27% for the present study, which is still below those of previous studies. Another factor that could explain a portion of the remaining difference is that of product mix. Kitchen and laundry appliances, which had low fulfillment rates, comprised roughly half of the durable plans in this study. However, confirmation of this potential explanation would necessitate detailed comparisons of the durables sets utilized in the various studies.

The percent of purchases that were unplanned varied greatly across product categories. Unplanned purchases were highest for kitchen appliances (59%) followed by entertainment appliances (48; ) and lowest for vehicles (19%). Laundry appliances also showed a very low percentage of unplanned purchases (20%) even though this product category had a low fulfillment rate. One might speculate that while consumers plan their laundry appliance purchases, these plans are relatively easy to postpone.

The percentage of purchases unplanned for first automobiles (18%) is substantially lower than figures found in previous work. McNeil (1974) found that buyers with no 12 month purchase plans accounted for 47.5% and 40% of 6 month car purchases reported in Census Bureau studies in the periods 1960-1966 and 1967-1972, respectively.

Planned vs. Unplanned Purchases

Earlier researchers have shown that the average length of the planning period varies considerably across individual durable product categories (Pratt 1968) and have concluded that there is no particularly strong relationship between consumers' forecasts of when a planned purchase will be made and the timing of actual purchase (Pickering and Isherwood 1974). Accordingly, the distribution of respondents in longitudinal studies such as ours between those making "planned" (anticipated and reported on the initial questionnaire) and "unplanned" purchases (not reported on the initial questionnaire) is more an artifact of the timing of the two questionnaire waves than a behaviorally meaningful division. Furthermore, it has been argued that most "unfulfilled plans" are merely postponed, not canceled, and many "unplanned purchases" reflect a prior policy or latent tendency to act, "deriving from expectations that are just as real but less specific than definite plans" (Foote 1973). For these reasons, we did not expect to find much difference in the main variables of interest in our analysis number of stores visited, number of brands considered, certainty at time of purchase that the best brand or model had been selected, and satisfaction with the item purchased between respondents making planned purchases and those whose purchases were classified as unplanned. These expectations of "no difference" are not inconsistent with earlier conceptual and normative writings that assume that planning before major purchases represents behavior that is somehow more rational and deliberate than impulsive purchases made without plans. Rather, our expectations here merely reflect the arbitrariness of classifying purchases as "planned" or "unplanned" based on two-wave survey findings. Nevertheless, we tested for possible differences between planned and unplanned purchases on several variables.

As expected, there was no statistically significant (p > .05) difference between planned and unplanned purchases in average number of brands considered, but those making planned purchases visited significantly (p < .05) more stores (x = 2.82) than did those making unplanned purchases (x = 2.37). However, when vehicles (which constituted a far more substantial segment of planned as compared to unplanned purchases) were removed from the analysis, this difference in stores visited disappeared. No statistically significant differences were found in comparing the distributions of the two groups' responses to the certainty and satisfaction questions. We further found that the distributions of payment method used did not vary between planned and unplanned purchases, even though it might be thought that the proportion of payment methods involving credit would be higher for unplanned than for planned purchases, in that credit availability might facilitate "impulsive" buying.

Four Acquisition Categories

Since the questionnaire item on acquisition categories I was included only for purchase plans, unplanned purchases I could not be classified by acquisition categories. Instead, respondents indicated whether each unplanned purchase was a replacement, ant, if so, whether the durable to be replaced was unusable and was to be discarded. As 6 a result, it was possible to determine whether unplanned purchases were Maintenance Replacements, Adjusting/ Upgrading Replacements, or those falling in the combined category of Additional Unit Expansion and First Acquisition Expansion. The relative importance of the acquisition categories appears to vary between unplanned and planned purchases and, to a lesser extent, between household items and vehicles, as Table 4 reveals. A substantial number of vehicle purchases are planned and represent replacements of the adjustment/upgrading type. A comparison of planned and unplanned purchases excluding vehicles shows the somewhat surprising result that maintenance replacement purchasers were relatively more common among planned purchases (38.3%) than among unplanned purchases (15.5%). Panel members were apparently very successful in anticipating needed replacements.

Results of tests of our major hypothesis are summarized in Table 5. For each of the five variables across the table, two cross-classifications were run with the four acquisition types, once with vehicles included and once without. Analysis of variance was conducted for all variables except Payment Method. These analyses were run only on the planned purchases in order to utilize the full acquisition category scheme. The interpretation of the significance measures depends somewhat upon whether the ANOVA or the cross-classification is considered more relevant, but several significant relationships are suggested by the results. These results provide at least partial support for our general expectation that the four acquisition categories would provide a better explanation for variability in search, certainty and satisfaction than the distinction between planned and unplanned purchases. As Table 5 shows, Maintenance Replacement purchases involved the least information search of the four acquisition categories and were most likely to be financed by one of the payment methods classified as Cash. Additional Unit Expansion purchases involved the most information search and resulted in the least satisfaction.

TABLE 2

PLANS AND PURCHASES

TABLE 3

FULFILLMENT RATES FOR SEVERAL MAJOR STUDIES OF TWELVE-MONTH DURABLE GOODS PURCHASE PLANS

TABLE 4

DISTRIBUTIONS OF PLANNED AND UNPLANNED PURCHASES FOR FOUR ACQUISITION CATEGORIES

TABLE 5

STORES VISITED, BRANDS CONSIDERED, CERTAINTY, SATISFACTION, AND PAYMENT METHOD BY FOUR ACQUISITION CATEGORIES OF PLANNED PURCHASES

Table 6 presents a summary comparison of the measures of association connected with the ANOVA (eta) and chi square (contingency coefficient) analyses using the two classification schemes. Although the acquisition categories are not highly related to the five dependent measures (i.e., number of stores visited, number of brands considered, degree of certainty, level of satisfaction, and payment method), this method of classifying purchases consistently outperforms the traditional planning scheme. However, it must be noted that these measures are not completely comparable since the acquisition analyses were performed only on planned purchases and the contingency coefficient is affected by the number of categories utilized.

Do Certainty and Satisfaction Increase with Search?

Correlations run between each of the two search measures (number of stores visited and number of brands considered) and the two "outcome" measures (certainty and satisfaction) provided only partial support for the intuitively plausible notion that purchasers whose search process included more stores and brands would be more certain at the time of purchase that the brand selected was the most appropriate for the need and would be more satisfied later with the performance of the item.

TABLE 6

A SUMMARY COMPARISON OF THE EXPLANATORY ABILITIES OF THE ACQUISITION AND PLANNING SCHEMES

While certainty was higher when more stores were visited (r = .141, p = .031), this relationship apparently was influenced largely by vehicle purchases. With vehicles removed from the analysis, r dropped to .045. However, certainty was correlated at a highly significant level with number of brands considered both for all purchases (r = .181, p = .005) and with vehicles removed from the calculation (r = .177, p = .02).

Correlations between the two search measures and satisfaction were all very low (r ranged from .03 to .09) whether or not vehicles were included in the calculation. Thus, certainty increased with number of brands considered and with number of stores only for visited or number of brands considered for any category. Not surprisingly, correlations between satisfaction and certainty for each product category and for all products together were very low (r ranged from .056 to .177) and were not significant.

CONCLUSIONS

Although we are puzzled by our panel's low fulfillment rate, which is lower than that found in previous studies, we are very encouraged by the finding that the four acquisition categories defined in the conceptualization of the study seemed to account for at least some of the variability in our main variables of interest--search, satisfaction, and payment method. At the same time, our findings were perfectly consistent with the notion that the distinction between "planned" and "unplanned" purchases is a methodological artifact. We feel that we should not attempt detailed interpretation of these acquisition categories and our findings at this point. However, additional measures available in our data set will provide the opportunity to further explore their meaning.

Our panel members reported low search and reasonably high satisfaction with their purchases. This finding may be puzzling to some, but it is not necessarily inconsistent with earlier findings that the relationship between search and satisfaction may be curvilinear (Westbrook and Fornell 1979). Certainty at time of purchase with respect to the suitability of the brand, not surprisingly, was related to the number of brands considered. Worthy of further investigation, though, is the finding that certainty and satisfaction are not related. The satisfaction measures may not be comparable with those obtained in other research because the ownership period for these products was very short.

REFERENCES

Alderson, Wroe (1957), Marketing Behavior and Executive Action, Homewood, IL: Richard D. Irwin, Inc.

Dickson, Peter R. and Wilkie, William L. (1978), The Consumption of Household Durables: A Behavioral Review, Cambridge: Marketing Science Institute.

Ferber, Robert (1954), "The Role of Planning in Consumer Purchases of Durable Goods," American Economic Review, 44 854-874.

Foote, Nelson (1974), "Unfulfilled Plans and Unplanned Actions," in Scott Ward and Peter Wright (eds.), Advances in Consumer Research, Vol. 1, Urbana: Association for Consumer Research, 529-531.

Gabor, Andre and Granger, C. W. J. (1972), "Ownership and Acquisition of Consumer Durables: Report on the Nottingham Consumer Durables Project," European Journal of Marketing, 6, 234-248.

Granbois, Donald H. (1977), "Consumer Shopping Behavior and Preferences," in Robert Ferber (ed.), A Synthesis of Selected Aspects of Consumer Behavior, Washington: U.S. Government Printing Office for the National Science Foundation. 259-298.

Granbois, Donald H. and Braden, Patricia L.(1976), "'Good Consumership' in Household Appliance Purchasing," Journal of Business Research, 4, 103-116.

Granbois, Donald H. and Summers, John O. (1975), "Primary and Secondary Validity of Consumer Purchase Probabilities," Journal of Consumer Research, 1, 31-38.

Granbois, Donald H. and Willett, Ronald P. (1968), "An Empirical Test of Probabilistic Intentions and Preference Models for Consumer Durables Purchasing," in Robert King (ed.), Marketing and the New Science of Planning, Chicago: American Marketing Association, 401-408.

Hill, Reuben (1963), "Judgment and Consumership in the Management of Family Resources," Sociology and Social Research, 47, 424-437.

McNeil, John (1974), "Federal Programs to Measure Consumer Purchase Expectations, 1946-1973: A PostMortem," Journal of Consumer Research, 1, 1-10.

Newman, Joseph W. (1977), "Consumer External Search: Amount and Determinants," in Arch Woodside et al., (eds.), Consumer and Organizational Buying Behavior, New York: Elsevier-North holland, 79-94.

Olshavsky, Richard W. and Granbois, Donald H. (1979), "Consumer Decision Making: Fact or Fiction?" Journal of Consumer Research, 6, 93-100.

Pickering, J. F. and Isherwood, B. C. (1974), "Purchase Probabilities and Consumer Durable Buying Behavior," Journal of the Market Research Society, 16, 203-226.

Pratt, Robert W. Jr. (1968), "Consumer Buying Intentions as an Aid in Formulating Marketing Strategy," in Robert King (ed.), Marketing and the New Science of Planning, Chicago: American Marketing Association, 296-302.

Theil, H. and Kosobod, R. F. (1968), "How Informative are Consumer Buying Intentions Surveys?" Review of Economics and Statistics, 50.

Westbrook, Robert A. and Fornell, Claes (1979), "Patterns of Information Source Usage Among Durable Goods Buyers,' Journal of Marketing Research, 16, 303-312.

----------------------------------------